[R] likelihood for new data from fitted nlme object

From: Andrew Bateman <awb42_at_cam.ac.uk>
Date: Tue, 12 Apr 2011 12:50:52 +0100

Dear list,

I have a fitted nlme object from which I want to produce estimates of the (marginal) likelihood for new data, given the fitted model.  I am trying to cross validate a number of nonlinear mixed effects models, and I would like to find a way to calculate the marginal likelihood for "virgin" data quickly.

Since nlme must approximated the marginal likelihood many times when fitting a model, I feel like there should be an easy way to do this for new data.

Is there an easy way to get this using the nlme model object?  Is there a straightforward way to manually generate estimates using components of the fitted model objects (e.g. using equation 4.2 from Lindstrom and Bates1990 "Nonlinear Mixed Effects Models for Repeated Measures Data")?

Many thanks (and apologies if I have failed to conform to any of the list guidelines),

Andrew Bateman
PhD Candidate
Department of Zoology
University of Cambridge

R-help_at_r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Tue 12 Apr 2011 - 12:34:59 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Tue 12 Apr 2011 - 12:50:29 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive